Deep sentiments in Roman Urdu text using Recurrent Convolutional Neural Network model

作者：

Highlights：

• This study contributes to generating the Roman Urdu corpus by annotating 10,021 reviews for the task of sentiment analysis.

• The deep learning models such as RCNN can help in text classification task for the under resource Urdu language.

• The RCNN outperforms Rule-based and N-gram methods, due to its ability to learn a language-independent textual context.

摘要

•This study contributes to generating the Roman Urdu corpus by annotating 10,021 reviews for the task of sentiment analysis.•The deep learning models such as RCNN can help in text classification task for the under resource Urdu language.•The RCNN outperforms Rule-based and N-gram methods, due to its ability to learn a language-independent textual context.

论文关键词：Roman Urdu corpus,Urdu sentiment analysis,Under resource language,Deep learning,Text Classification,Recurrent Convolutional Neural Networks (RCNN)

论文评审过程：Received 10 August 2019, Revised 20 February 2020, Accepted 27 February 2020, Available online 6 March 2020, Version of Record 6 March 2020.

论文官网地址：https://doi.org/10.1016/j.ipm.2020.102233